Daily Self-Compassion and Preservice Teachers’ Professional Identity Development: A 14-Day Intensive Longitudinal Study

Technical Summary: Daily Self-Compassion and Preservice Teachers' Professional Identity Development

Problem Statement
Teacher professional identity is a dynamic, situated, and emotional construct central to preservice teachers' commitment and pedagogical decision-making. While existing literature establishes that self-compassion is a beneficial psychological resource for university students, most research on the link between self-compassion and professional identity in teacher education relies on cross-sectional or broad longitudinal designs. These approaches identify stable between-person correlates but fail to distinguish whether daily fluctuations in self-compassion co-occur with daily fluctuations in professional identity (within-person effects) or whether self-compassion on one day predicts identity changes the following day (lagged effects). This study addresses the need to disentangle stable trait-like differences from day-to-day state-like processes within the specific context of Chinese preservice teacher education, where high academic and institutional pressures may make emotional regulation critical.

Methodology
The study employed a 14-day intensive longitudinal diary design with 267 Chinese preservice teachers (146 in preschool education, 121 in primary education) from Shaanxi Xueqian Normal University.

• Data Collection: Participants completed a daily questionnaire for 14 consecutive days, yielding 3,473 valid observations (92.9% completion rate).
• Measures:
  • Self-Compassion: Assessed using state-adapted items from the Self-Compassion Scale-Short Form. Item diagnostics revealed that two of the six original items (Items 5 and 6) showed anomalous patterns. Consequently, a four-item core self-compassion score was established as the primary measure, while the six-item raw and recoded versions were retained for sensitivity analyses.
  • Professional Identity: Assessed using four newly developed state-adapted items reflecting perceived fit, value identification, willingness to engage, and positivity toward future career direction.
• Analytic Strategy: The analysis utilized multilevel modeling and observed-score multilevel dynamic regression (following DSEM principles) to separate between-person (stable) and within-person (daily) variance.
  • Reliability: Multilevel reliability estimates were calculated to justify the scoring strategy.
  • Associations: The study examined between-person correlations, within-person same-day associations, and lagged (cross-lagged) associations using a bivariate multilevel VAR(1) model.
  • Model Comparison: Nested models (autoregressive-only, cross-lagged, and partial contemporaneous) were compared using AIC, BIC, and $R^2$ to determine the best-fitting dynamic specification.
  • Heterogeneity: An exploratory random-slope mixed model tested whether the strength of the same-day association varied across participants.

Key Results

1. Measurement Validity: The six-item raw self-compassion score demonstrated poor reliability ($\alpha_{raw} = .340$) due to inconsistent item functioning. The four-item core score showed adequate reliability ($\alpha_{within} = .753$, $\alpha_{between} = .950$) and was used for primary analyses. Sensitivity analyses confirmed that substantive conclusions held across all scoring approaches.
2. Between-Person Association: Preservice teachers with higher average self-compassion reported stronger average professional identity ($r = .414, p < .001$). This association remained significant after controlling for major.
3. Within-Person Same-Day Association: On days when participants were more self-compassionate than their own average, they reported stronger professional identity than usual ($r = .268, p < .001$). In a partial contemporaneous model controlling for prior-day values and time, current-day self-compassion significantly predicted current-day professional identity ($\beta = .263, p < .001$).
4. Lagged Dynamics: The study found no significant evidence that prior-day self-compassion predicted next-day professional identity ($\beta = .023, p = .341$) after controlling for prior-day identity. Conversely, prior-day professional identity did predict current-day identity ($\beta = .100, p = .002$), indicating autoregressive stability.
5. Model Fit: Model comparison indices favored the partial contemporaneous model over the cross-lagged model (Delta AIC = -221.62), supporting the conclusion that the relationship is primarily simultaneous rather than a one-day carryover effect.
6. Individual Differences: A random-slope model indicated significant heterogeneity in the same-day effect ($\text{variance} = .102$), suggesting that the coupling between self-compassion and identity varies meaningfully across individuals.

Significance and Claims
The paper claims that self-compassion functions primarily as a same-day psychological resource for preservice teachers rather than a robust one-day driver of identity change.

• Theoretical Contribution: The findings support a "daily resource perspective," suggesting that professional identity is both anchored in stable commitments and flexible enough to shift with proximal daily appraisals. The study clarifies that while self-compassion is linked to identity, this link operates through contemporaneous appraisal processes (how one treats oneself today affects how one feels about the profession today) rather than through a simple temporal lag.
• Contextual Relevance: In the Chinese teacher-education context, characterized by high expectations and strict evaluation, self-compassion may help students interpret difficulties as part of shared professional growth rather than evidence of unsuitability.
• Methodological Rigor: The study highlights the necessity of intensive longitudinal methods to distinguish between-person traits from within-person states. It also demonstrates the importance of measurement transparency, showing that raw scale scores can be psychometrically flawed in daily diary contexts and require item-level diagnostics.
• Practical Implication: The authors suggest that interventions should focus on embedding self-compassion practices close to stressful learning episodes (e.g., after feedback or practicum challenges) rather than assuming a single intervention will have delayed, lasting effects on identity. The variability in individual effects further cautions against one-size-fits-all approaches.

The authors conclude that while self-compassion is a meaningful resource for "professional becoming," its influence is immediate and daily, requiring ongoing engagement rather than a one-time shift in mindset.